GPU & ML Developer Reconstruction and Simulation; EP-ALI-SC--GRAP

ALICE is pioneering the use of GPUs in Run 3 for the online processing and partly for offline reconstruction. To better leverage available GPU compute resources and improve reconstruction performance, we aim to investigate the use of machine learning.

As a GPU and ML software developer, you will maintain, develop, and commission machine-learning-based GPU event reconstruction code for the ALICE experiment, in particular ML-based and ML-supported clusterisation, and track seeding in the ALICE TPC.

In parallel, you will contribute to ALICE's Monte Carlo production ecosystem and simulation frameworks, focusing on workflow optimisation. This includes the full MC production infrastructure, simulation frameworks, automation of production, validation and integration of ML and GPU-code, and the development and use of intelligent computing tools across the ALICE computing chain.

• Commission the GPU TPC ML clusterisation as the default clusterisation code for data taking and for simulation.
• Benchmark and improve the ML-based clusterisation in terms of processing performance and physics quality.
• Investigate extending ML usage, including to TPC track seeding.
• Contribute to the Monte Carlo production ecosystem, including workflow scheduling, multi-timeframe processing, multi-threading, and integration of ML/GPU components.
• Develop and operate automated solutions for MC production, job orchestration, and validation, including ML-based anomaly detection.
• Track the activities in the optimisation and modernisation of simulation and reconstruction frameworks (e.g. Geant, Alice
  
  O2), including ML-driven acceleration and GPU-based approaches.
• Investigate components and algorithms of the ALICE computing chain (simulation, reconstruction, etc.) that could benefit from machine learning and develop prototypes.

• Experience with high energy physics (HEP) experiments event reconstruction code (e.g. clusterisation or tracking).
• Experience with GPU programming and ML training and inference.
• Practical experience with debugging large distributed applications.

• Strong knowledge of the C++ programming language on Linux.
• Knowledge of at least one GPU programming toolkit such as CUDA or HIP.
• Knowledge of an ML framework such as ONNXRuntime.
• Knowledge of debugging tools such as GDB and profiling tools such as perf.
• Ability to work in a team.
• Spoken and written English, with a commitment to learn French.

• You are a national of a CERN Member or Associate Member State.
• You have a professional background in Physics (or a related field) and have either:
  • a Master's degree with 2 to 6 years of post-graduation professional experience;
  • or a PhD with no more than 3 years of post-graduation professional experience.
• You have never had a CERN fellow or graduate contract before.

Job closing date:
01.07.2026 at 23:59 CEST.

Contract duration: 24 months, with a possible extension up to 36 months maximum.

Working hours:

40 hours per week

Job flexibility:
Fully Onsite

Target start date: 01-August-2026

• Participation in a regular stand‑by duty, including nights, Sundays and official holidays.
• Stand‑by duty, when required by the needs of the Organization.

Job reference: EP-ALI-SC--GRAP

Field of work:
Applied Physics

Benchmark job: 200140 - Applied Physicist

• A monthly stipend between  Swiss Francs per month (tax free) depending on your degree.
• 30 days of paid leave per year plus 2 weeks annual closure.
• Coverage by CERN’s comprehensive health insurance scheme (for yourself, your spouse and children), and membership of the CERN Pension Fund.
• Family, child and infant monthly allowances depending on your individual circumstances.
• A relocation package (installation grant and travel expenses) depending on your individual circumstances.
• Possibility to extend your contract up to 36 months.
• On-the-job and formal training including language classes.